unexploitable nl play

[tmcdmck]

is it possible to play unexploitable poker with 100 bb stacks? obviously being a better player in every way would work, but i was wondering if there was something equivalent to the SAGE system (although obviously infinitely more complex). I realise this method of play is probably practically impossible to work out, and probably sub optimal against a skilled player, but does anyone reckon such is system exists?

[jstill]

heres the biggest mistake people make about thinking about playing unexploitable poker (mainly this thought derives from my limit play/ thinking but it applies to NL as well): U dont need to concern yourself with playing completely optimal unless your opponents are capable of exploiting unbalanced play and adjust very quickly. Most players aren't thinking about your range and what you're liable to do with each hand and how they can exploit that anyways, most players (and I think this is even more try in small stakes NL than limit) are playing almost exclusively 1st level poker and even when they go second or (rarely) third level, they do so poorly. So thats where you're wrong, playing optimal unexploitable poker is the most useful vs SKILLED players, not the other way around. VERY IMPORTANT.

Ideally you want to play as exploitable (unbalanced) of a game as you can get away with and just focus on making the best decisions in each instance and not concerning yourself about your overall strategy (exploitability) unless you commonly see the same players (and they are capable of making adjustments exploiting you) ie they play well (if they play poor, but adjust to your consistent play just readjust, you dont need to think in terms of overall unexploitable play, just exploit their weaknesses to the max and recognize changes in their tendencies vs you over time). Most of our opponents are idiots (ideally with table selection), so playing unexploitable poker vs them is not "optimal" in the terms of it being what will net you the most money.

[A.Nironen]

[ QUOTE ]
I realise this method of play is probably practically impossible to work out, and probably sub optimal against a skilled player, but does anyone reckon such is system exists?

[/ QUOTE ]

Yes, it exists. It was proven by John Forbes Nash Jr. in 1950.

Andrzej

[Hi5]

It exists for HU play. For other games, you may have new problems, such as implicit collusion.

[aszmel]

jstill, deeply think like you, on lower stakes just focusing on exploiting opponents tendencies and their holes is most important, if opponent is wise and playing back at our hole (due to opening it to exploit others) we should close it and found another leaks at their strategies, but isn't it wise to first know how to play optimal and unexploitable to know how to go back from exploitable way to unexploitable one? to know how much we went away? of course on low stakes it's not so important, but learning process should be done on lower stakes before (hopefully) we hit higher stakes...

[Josem]

[ QUOTE ]
is it possible to play unexploitable poker with 100 bb stacks? obviously being a better player in every way would work, but i was wondering if there was something equivalent to the SAGE system (although obviously infinitely more complex). I realise this method of play is probably practically impossible to work out, and probably sub optimal against a skilled player, but does anyone reckon such is system exists?

[/ QUOTE ]

whether or not an unexploitable play exists is a little irrelevant.


unexploitable play hasn't even been found for chess yet - and that's a game where 100% of the knowledge is available.


the idea of finding such a system for poker - where there is the added randomness of cards and unknown information - is on such a scale of magnitude so much harder that it isn't even feasible to consider the possibility yet.



i suspect such a sytem is theoretically possible, but I can't imagine it being developed within the near future.

[mvdgaag]

Any game that allows mixed strategies has an optimal strategy, so yes, it exists.
As others have pointed out it is a holy grail not worth looking for. Optimal play does a terrible job maximising profit in about any game you can find on earth. I don't know about high stakes though. It is good to know what strategies are suboptimal and how to exploit them, because thats how you make a lot more money than playing optimal.

[helter skelter]

[ QUOTE ]

the idea of finding such a system for poker - where there is the added randomness of cards and unknown information - is on such a scale of magnitude so much harder that it isn't even feasible to consider the possibility yet.

i suspect such a sytem is theoretically possible, but I can't imagine it being developed within the near future.

[/ QUOTE ]

That may be true for hold-em but not for all forms of poker. Game theory has already been developed for draw poker and probabaly could be (or maybe has been) for similar games where there are only 2 rounds of betting.

I'm not sure if having exposed cards makes it more complex, or if it is just because of the multiple rounds of betting, but with stud and hold-em, I agree that even if possible, it would probably take a mathemetician a lifetime to develop such a strategy, and then you would need to be an android like Data to use it in practice, able to draw on an encyclopedic memory and do instantaneous calculations for each betting round.

[ev_slave]

[ QUOTE ]
unexploitable play hasn't even been found for chess yet - and that's a game where 100% of the knowledge is available.

[/ QUOTE ]

Ah, the point I was planning to make. Except yours was probably more concise.

But to add to the discussion, Chess is also a game where the number of possible moves are reasonably small. In poker, you have a much greater range of plays available. In a 5/10 game the smallest increment is a $5 chip (online, it's even smaller). So you decide to bet the flop with $30 in the pot... great! Do you bet $25? $30? 35? All the way up to your whole stack? Which is optimal?

This makes an optimal strategy just seem like a headache.

[tarheeljks]

[ QUOTE ]
Any game that allows mixed strategies has an optimal strategy, so yes, it exists.

[/ QUOTE ]

my game theory is rusty, but is this true in games of imperfect information?